Microsoft Models
Microsoft logo

Phi 4

by Microsoft

[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion parameters, it was trained on a mix of high-quality synthetic datasets, data from curated websites, and academic materials. It has undergone careful improvement to follow instructions accurately and maintain strong safety standards. It works best with English language inputs. For more information, please see [Phi-4 Technical Report](https://arxiv.org/pdf/2412.08905)

Chat with Phi 4
Input Price$0.13/1M tokens
Output Price$0.50/1M tokens
Intelligence10.5
Coding11.2

Specifications

Technical details and pricing.

ProviderMicrosoft
Context Window16,384 tokens
Release DateDec 12, 2024
ModalitiesText

Benchmarks

12 benchmark scores from Artificial Analysis.

GPQA57.5%
MMLU Pro71.4%
HLE4.1%
LiveCodeBench23.1%
MATH 50081.0%
AIME 202518.0%
AIME14.3%
SciCode26.0%
LCR0.0%
IFBench23.5%
Tau20.0%
TerminalBench Hard3.8%

Composite Indices

Intelligence, Coding, Math

Standard Benchmarks

Academic and industry benchmarks

Frequently Asked Questions

What is Phi 4 good for?

Use Phi 4 for everyday tasks like writing, summarizing, brainstorming, and getting clear explanations.

How much does Phi 4 cost?

Pricing is based on usage. Current rates are $0.13/1M tokens for input and $0.50/1M tokens for output.

Can I try Phi 4 for free?

Yes. You can start a chat instantly and test the model before deciding on a plan.

Does Phi 4 support images or audio?

Phi 4 focuses on text-based tasks.

Benchmarks and pricing are sourced from Artificial Analysis where available. OpenRouter specs are used as a fallback.